Monitoring device using selective attention model and method for monitoring same

Abstract

A monitoring device is provided, which includes an inputter configured to receive an input of a plurality of images captured at separate positions and a plurality of sound sources heard at separate positions, a saliency map generator configured to generate a plurality of mono saliency maps for the plurality of images and to generate a dynamic saliency map using the plurality of mono saliency maps generated, a position determinator configured to determine the positions of the sound sources through analysis of the plurality of sound sources, a scan path recognizer configured to generate scan paths of the plurality of images based on the generated dynamic saliency map and the determined positions of the sound sources, and an outputter configured to output the generated scan paths.

Description

TECHNICAL FIELD[0001]The present disclosure relates to a monitoring device and a monitoring method, and more particularly to a monitoring device using a selective attention model based on audio-visual integrated information and a monitoring method therein.BACKGROUND ART[0002]A monitoring device performs a function of accurately identifying an input image and sound using sensor technology. The sensor technology that is used in the monitoring device starts from imitation of human sensory organs. Recently, the importance of an active artificial visual system and voice recognition system has been highlighted as the sensor technology.[0003]However, many artificial visual systems that have been developed up to now lay stress on detection and recognition of a specific object in an input image, and thus do not have a solution to the problem in a start stage of the visual system, such as how to effectively select, like a human visual system, necessary information from a natural image of a re...

AI Technical Summary

Benefits of technology

The monitoring device and monitoring method described in this patent allows for the simultaneous analysis of video and audio information to create a scan path that mimics the natural movements of humans. This results in a high level of reliability. Additionally, the device can detect the time points when a person speaks, breaks down the audio into chunks, and selects the sound source for voice recognition.

Problems solved by technology

However, many artificial visual systems that have been developed up to now lay stress on detection and recognition of a specific object in an input image, and thus do not have a solution to the problem in a start stage of the visual system, such as how to effectively select, like a human visual system, necessary information from a natural image of a real world environment having high complexity.

Further, the existing voice recognition system has the problem that, since much noise exists in the real world environment, it is difficult to determine when to perform the voice recognition using only a voice signal, and it occurs frequently to execute the voice recognition by the noise rather than a speaker.

Images